The trajectory of increasing model complexity seems quite natural to me. There were global models and then regional ones; slab oceans and then stratified ones; low resolution and then higher resolution; and who knows what internal changes in the science and the parameters.

On the computer side there has not been a compensating application of Moore’s Law. Computers are faster but not fast enough; disks are certainly larger and I would guess large enough; memory is cheaper but oddly constrained by operating system and supplier (for Windows on my Dell, at least);

Really? I just built a Ryzen 9 3900XT. I put 64GB into it and it will take 128GB. Boinc needs nothing like that.

My point was intended to concern the failure of the speed of PCs to increase "as expected", though I concede that Moore's Law isn't really about speed. The number of processors has indeed increased and that means fewer users are needed for a fixed ensemble size, or if the same number of users carry on with the project then the ensemble sizes could be increased (if that is scientifically interesting). However, ensemble completion times won't have improved proportionally.

Here's a normalised estimate of the speed of the machines on which I have run CPDN models plotted against date of purchase, based on 653 models benchmarked across eight machines. The linear fit has a gradient of ~10%, which is a long way short of a speed version of Moore's Law. That's all I meant to say.

An interesting perspective Ian. I have had rather fewer machines since I started with the project and have relatively recently made the jumps from a core2 duo to a ryzen7 which means that as well as being able to run more tasks at once, they are completing in less than a third of the time if I run eight at once. a bit faster if I keep it down to five so over twelve times the throughput. Just a shame more machines aren't running Linux as the current flow of batches means it takes quite a long time to get through them.

I have noticed that the performance of laptops (the only kind of PC’s I have personal experience with) hasn’t increased much in the last few years. Going back 12 to15 years, it used to be that every time I replaced a machine (about every 2.5 to 3 years) the clock speed of the chip and amount of RAM were increased for the same money. Recently, this has not been true. Now if you want an increase in speed or RAM the price goes up at more than just the rate of inflation.

---

I have noticed that the performance of laptops (the only kind of PC’s I have personal experience with) hasn’t increased much in the last few years. Going back 12 to15 years, it used to be that every time I replaced a machine (about every 2.5 to 3 years) the clock speed of the chip and amount of RAM were increased for the same money. Recently, this has not been true. Now if you want an increase in speed or RAM the price goes up at more than just the rate of inflation.

The desktop CPUs are way faster than they used to be. Is it just laptops that haven't progressed? I can't see why there would be a difference.

The speed difference is probably down to a few parameters to do purely with laptops. The driving forces in this area have been weight and battery life as well as better screen resolution. Since battery life is dependant on size then compromises have to be made with the power that any chips can consume. Therefore the CPUs are lower power consumption and as a knock on effect from that slower or with lower overall computing capacity.

The speed difference is probably down to a few parameters to do purely with laptops. The driving forces in this area have been weight and battery life as well as better screen resolution. Since battery life is dependant on size then compromises have to be made with the power that any chips can consume. Therefore the CPUs are lower power consumption and as a knock on effect from that slower or with lower overall computing capacity.

So can you actually buy a laptop you can use on your lap now? The last one I used would overheat instantly, since the intake vent was underneath, where your lap is blocking it! It should have been sold as a tabletop.

And I thought a lot of gamers were going for laptops now, so surely they're adding a lot of power for that, or is that all in the GPU?

can you actually buy a laptop you can use on your lap now? The last one I used would overheat instantly, since the intake vent was underneath, where your lap is blocking it! It should have been sold as a tabletop.

And I thou

ght a lot of gamers were going for laptops now, so surely they're adding a lot of power for that, or is that all in the GPU?

Most if running flat out will over heat if sitting on a lap unless you use a tray between the laptop and lap. They are fine for just dealing with emails if air intakes are blocked.

I don't have a working laptop at the moment but wouldn't use one for anything too intensive as the fans get noisy. Liquid cooling means if listening to anything on speakers, I can't hear the fans on my desktop, even with 8 cores on full load. I don't run anything that is intensive on the graphics and have never played any demanding games ona computer so can't answer the last bit.

So can you actually buy a laptop you can use on your lap now? The last one I used would overheat instantly, since the intake vent was underneath, where your lap is blocking it! It should have been sold as a tabletop.

And I thought a lot of gamers were going for laptops now, so surely they're adding a lot of power for that, or is that all in the GPU?

Yeah, they do say not to hold a laptop on your lap because of overheating. The name has stuck from the old days. I think the proper term is notebook.

You can get faster gaming laptops if you are prepared to shell out big bucks. They are in the $1200 USD range. That is out of my price range. Recently, I have been buying “refurbished” laptops in the $350 USD range. These models usually sold originally for about $650 USD.

---

So can you actually buy a laptop you can use on your lap now? The last one I used would overheat instantly, since the intake vent was underneath, where your lap is blocking it! It should have been sold as a tabletop.

And I thought a lot of gamers were going for laptops now, so surely they're adding a lot of power for that, or is that all in the GPU?

Yeah, they do say not to hold a laptop on your lap because of overheating. The name has stuck from the old days. I think the proper term is notebook.

You can get faster gaming laptops if you are prepared to shell out big bucks. They are in the $1200 USD range. That is out of my price range. Recently, I have been buying “refurbished” laptops in the $350 USD range. These models usually sold originally for about $650 USD.

Surely, especially a gaming laptop, could just have the intake and exhaust both on the edge? A very simple and obvious thing to do. Stupid designers.... People put laptops on their lap when sat on the sofa, or they'd use a desktop!

There's something like 400 CPU-years of work queued.

Yes, and all of it is set to run only on Os’s that are run by only a tiny minority of the public. Windows is run on about 87% of all the home computers in the World.[/quote]

I personally don't have a problem with that, Linux OS's can be burnt onto a dvd and up and running in under 15 minutes, but that's beside my point...my point is I can't find WHERE it says that X type of task is only for 32bit Linux OS's while Y type of task is for 64bit Linux OS's and Z type of task over there is for Windows 7 or 10 pc's. A bit more info instead of guessing would help especially since the default timeout of 3600 seconds makes any kind of figuring out what kind of task will work on my pc useless.

A bit more info

That is a 2 step process:
1. Look at the Project status page, at the Tasks by application section, in the Application column, to find the name of the application in which you're interested.
2. Look on the Applications page to find out which OS's have that Application.

**********************

And don't forget: the tasks are from research groups external to Oxford.
What OS each group uses, is entirely up to them.

There's something like 400 CPU-years of work queued.

Yes, and all of it is set to run only on Os’s that are run by only a tiny minority of the public. Windows is run on about 87% of all the home computers in the World.

I personally don't have a problem with that, Linux OS's can be burnt onto a dvd and up and running in under 15 minutes, but that's beside my point...my point is I can't find WHERE it says that X type of task is only for 32bit Linux OS's while Y type of task is for 64bit Linux OS's and Z type of task over there is for Windows 7 or 10 pc's. A bit more info instead of guessing would help especially since the default timeout of 3600 seconds makes any kind of figuring out what kind of task will work on my pc useless.[/quote]

How it is now -- Windows 10 has a subsystem in the windows store for free that makes it easy peasy to run Ubuntu and a few other Linux apps under Windows 10. I myself have tried that. It works reliably but I've not compared efficiency
for a long time Linux has had "not linux - independent FOSS project called "wine" " that has worked for me

Actually on Win 10 or linux it takes a small bit of work --

What I say is , with an hour reading the documentation,
Anybody can run "Windows under Linux" or "Linux under Windows"
Both "just work" -- takes an hour or two -- you can run most apps either way.

If anybody wants it easier than that -- Duh

e

---

quote]Windows is run on about 87% of all the home computers in the World.[/quote]

And another thing: climate research is run by people using big Unix / Linux systems, not Windows.
What the public uses to do things is irrelevant.

Windows is run on about 87% of all the home computers in the World.

And another thing: climate research is run by people using big Unix / Linux systems, not Windows.
What the public uses to do things is irrelevant.

It's relevant when you're outsourcing the calculations to people with Windows. And it may be fairly easy to run Linux stuff under Windows, but most people won't bother. I'm technically minded and have been in IT support for 11 years officially, and far longer unofficially. I could run the Linux tasks under Windows. But why should I? There are so many projects that have native Windows tasks available, I just do those instead. And most people probably wouldn't be able to or want to run Linux under Windows. Msot users don't know much about the techy stuff.

We're NOT outsourcing work to people with Windows.
It's being provided to people with appropiate computer systems.
And for about a year now, that has NOT been Windows people.

I expect the researchers do their own thing on their own schedules. They may have no idea what the other researchers are doing.
So it is not surprising that we can for years without any Linux work either, and then it dumps on us all at once.

That does not concern me so much as the lack of selectivity. We have to run all of it or nothing.

And most people probably wouldn't be able to or want to run Linux under Windows. Msot users don't know much about the techy stuff.

Personally, I got involved with CPDN because I believe in the importance of climate research. What you say is no doubt true for those who just want to collect credit. (Though, there are projects which give a higher return/cpu hour than CPDN.) I learned the relatively simple process of running Windows tasks under Linux during the many years when the overwhelming majority of work was for Windows.

The other side of it is that researchers with deadlines for their Masters/PhD who are familiar with Unix/Linux systems probably don't have the time to learn enough to port their applications to Windows, especially if it is the task type that runs well under Linux that is best suited to their research. I would guess they have enough other work to be getting on with while waiting the two or three months for enough results to come back when there aren't enough Linux machines to finish them more quickly.

And most people probably wouldn't be able to or want to run Linux under Windows. Msot users don't know much about the techy stuff.

Personally, I got involved with CPDN because I believe in the importance of climate research. What you say is no doubt true for those who just want to collect credit. (Though, there are projects which give a higher return/cpu hour than CPDN.) I learned the relatively simple process of running Windows tasks under Linux during the many years when the overwhelming majority of work was for Windows.

The other side of it is that researchers with deadlines for their Masters/PhD who are familiar with Unix/Linux systems probably don't have the time to learn enough to port their applications to Windows, especially if it is the task type that runs well under Linux that is best suited to their research. I would guess they have enough other work to be getting on with while waiting the two or three months for enough results to come back when there aren't enough Linux machines to finish them more quickly.

___________________________

Kudos to Dave for a concise answer.
Peter, for credits, compute "Collapsed Conjecture". We compute this project for the love of the subject. Climate. So, if love demands we run Linux on Virtual Box, so be it.

And most people probably wouldn't be able to or want to run Linux under Windows. Msot users don't know much about the techy stuff.

. You will be surprised how many are.

It took me about 45 minutes to set up a VM with 64bit Ubuntu Mate 20.04 LTS, installing Boinc, installing the needed libraries with

sudo apt install lib32ncurses6 lib32z1 lib32stdc++-7-dev

and add CPDN to my Boinc Manager. To be fair, i have installed Linux with a GUI, because i cannot work just with a console. But when you install Linux with an GUI, the navigation is nearly the same as in Windows. Yes, it depends on which Linux distribution you use, thats why i mentioned, that I use Ubuntu Mate.

Greets from Germany
Felix

Peter, for credits, compute "Collapsed Conjecture". We compute this project for the love of the subject. Climate. So, if love demands we run Linux on Virtual Box, so be it.

This isn't the only worthy project. And it isn't the only climate project. I'll just pick the ones that use Windows.

And most people probably wouldn't be able to or want to run Linux under Windows. Msot users don't know much about the techy stuff.

.You will be surprised how many are.

Most people can't even use windows. If you've ever worked in IT support, you'll know most computer users are like most car drivers, they have no idea how it works inside.